AI 기반 마커리스 동작 분석 시스템을 활용한 무용 숙련도별 보행의 시공간적 변인 비교 분석

Objectives The purpose of this study was to investigate differences in spatio-temporal gait parameters among dancers of varying skill levels using an AI-based markerless motion capture system. Methods Forty-five participants were classified into three groups: the expert-level (EX, n=15), the intermediate level (IM, n=15), and the untrained-control (UC, n=15). Participants performed overground gait trials along a 5-meter walkway at their preferred natural speed while barefoot. Gait data were captured and recorded using the KinaTrax AI markerless system. Statistical analysis was performed using SPSS 28.0 with one-way ANOVA followed by Tukey’s post-hoc tests (p < .05). Results Significant group differences were found in stride length, swing time, and gait speed. The EX group exhibited significantly greater overall stride length and longer swing time compared to both the IM and UC groups (p < .05). Specifically, the EX group showed a faster gait speed than the UC group (p = .03). In terms of gait asymmetry, all participants displayed excellent gait symmetry, and no statistically significant differences were observed in gait asymmetry among the three groups (p > .05). Conclusions Expert-level dancers exhibited distinct gait characteristics, specifically longer stride lengths, faster gait speeds, and extended swing times, suggesting enhanced dynamic balance and sophisticated motor control. These proficiency-related differences may reflect gradual physical adaptation to the duration and intensity of dance training. The findings of this study can provide a scientific foundation for developing evidence-based training programs and establishing tailored injury prevention guidelines that account for dancers' skill levels.